Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

The computation of electron repulsion integrals (ERIs) is the most time-consuming process in the density functional calculation using Gaussian basis set. Many temporal ERIs are calculated, and most are stored on slower storage, such as cache or memory, because of the shortage of registers, which are the fastest storage in a CPU. Moreover, the heavy register usage makes it difficult to launch many concurrent threads on a graphics processing unit (GPU) to hide latency.
Hence we propose to optimize the calculation order of one-center ERIs to minimize the number of registers used, and to calculate each ERI with three or six cooperating threads. The performance of this method is measured on a recent CPU and a GPU. The proposed approach is found to be efficient for high angular basis functions with a GPU. When combined with a recent GPU, it accelerates the computation almost 4-fold.

DOI： 10.1002/qua.24607

Language：English  

Authorship：Lead author   Language：Japanese  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

With the density-matrix variational method, the ground-state energy of a many-electron system is calculated by minimizing the energy expectation value subject to the selected representability conditions for the second-order reduced-density matrix (2-RDM). There is a lack of size extensivity under the P, Q, and G conditions of this method by applying to M noninteracting atoms and molecules up to M=32; BeM, (CH4)M, and (N2)M. The energy per molecule (E/M) monotonically decreases as M increases, showing the lack of size extensivity in this method. The inextensive contributions to energies are 3×10-4 and 3×10-3 a.u. for (CH4)M and (N2)M using the STO-6G basis set, respectively. In the examples studied, E/M approaches a finite value as M-->[infinity]. The intermolecular elements of 2-RDMs do not satisfy the cumulant expansion.

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

By using the many-particle Green's function (GF) the extension of the fragment molecular orbital (FMO) method by Kitaura et al. [Chem. Phys. Lett. 313, 701 (1999)] is proposed. It is shown that the partial summation of the cluster expansion of GF reproduces the same extrapolation formula as that of FMO. Therefore we can determine the excitation energy, the transition moment, and the linear response of a molecule from GF approximated with the FMO procedure. It is also shown that no wave function exists which is consistent to the FMO results. The perturbation expansion in which the self-consistent charge approximation defines the unperturbed state is reported. By using it the three-body effects missing in the pair approximation of FMO are analyzed and the corrections to the energy and the reduced density matrices are proposed. In contrast to the previous works these new corrections are not expressed as the addition or the subtraction of the energies of fragments. They are size extensive and require only the quantities available by the FMO calculation.

Language：English   Publishing type：Research paper (scientific journal)  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Language：English   Publishing type：Research paper (scientific journal)  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Language：English   Publishing type：Research paper (scientific journal)  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Authorship：Lead author   Language：English   Publishing type：Research paper (scientific journal)  

Event date： 2015.10

Language：English   Presentation type：Oral presentation (invited, special)  

Country：Slovakia  

Event date： 2015.6

Language：English   Presentation type：Poster presentation  

Country：Japan  

Event date： 2013.12

Language：English   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2013.6

Language：Japanese   Presentation type：Oral presentation (invited, special)  

Country：Japan  

Event date： 2013.4

Language：English   Presentation type：Poster presentation  

Country：United States  

Event date： 2012.12

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：自然科学研究機構　岡崎コンファレンスセンター   Country：Japan  

Event date： 2012.10

Language：English   Presentation type：Poster presentation  

Country：Japan  

The possibility of the molecular modeling and the virtual screening increases as the increase of the computational resources. Today biologically important macromolecules, such as proteins, are within the scope of electronic-structure calculation by using fragment molecular orbital (FMO) method. Also there are several well established methods to measure chemical similarity computationally. LINGO is such a method by comparing SMILES representations of two molecules. Looking at the development of recent computer, many-core processors such as graphics processing units (GPUs) would be key technology for future high-performance computation. General-purpose GPU (GPGPU) is successfully applied for many kinds of problems in high performance computation.
Since 2007 we developed the method to accelerate the molecular orbital calculation by using NVIDIA GPUs. This study produced the XA-CUDA-QM add-in library to accelerate standard QM software (GAMESS-US etc). Recently we extended its functionality such as using d-orbital basis set, FMO trimer calculation (FMO3), and energy gradient calculation, which will be presented. We also examine a performance on the newest NVIDIA's GPU, GTX680, called Kepler, and found it excellent to accelerate FMO.
Second, we develop a LINGO similarity calculation method to utilize both CPUs and GPUs at the same time. Our method is based on SIML reported to accelerate LINGO by using GPGPU.

Event date： 2012.9

Language：Japanese   Presentation type：Poster presentation  

Venue：東京大学　本郷キャンパス   Country：Japan  

Event date： 2012.3

Language：English   Presentation type：Poster presentation  

Country：Japan  

Event date： 2012.2

Language：English   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2011.12

Language：English   Presentation type：Poster presentation  

Country：Japan  

Event date： 2011.11

Language：Japanese   Presentation type：Oral presentation (keynote)  

Venue：長崎大学文教キャンパス   Country：Japan  

Event date： 2011.10

Language：English   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2011.9

Language：Japanese   Presentation type：Oral presentation (general)  

Venue：札幌コンベンションセンター   Country：Japan  

Event date： 2011.5

Language：English   Presentation type：Oral presentation (general)  

Country：Czech Republic  

Event date： 2011.5

Language：English   Presentation type：Poster presentation  

Country：Czech Republic  

Event date： 2011.3

Language：English   Presentation type：Poster presentation  

Country：Japan  

Event date： 2010.11

Language：English   Presentation type：Poster presentation  

Event date： 2010.11

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2010.9

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2010.9

Language：Japanese   Presentation type：Poster presentation  

Country：Japan  

Event date： 2010.9

Language：Japanese   Presentation type：Oral presentation (general)  

Country：Japan  

Event date： 2010.7

Language：Japanese   Presentation type：Oral presentation (invited, special)  

Event date： 2009.8

Language：English   Presentation type：Poster presentation  

Country：Japan  

Event date： 2009.5

Language：Japanese   Presentation type：Oral presentation (invited, special)  

Country：Japan  

Event date： 2001

Language：English   Presentation type：Poster presentation  

Event date： 2000

Language：English   Presentation type：Oral presentation (invited, special)  

Country：Japan  

物理化学、量子化学の入門を講義する。

数値計算、古典力学の計算機シミュレーション、並列処理について講義し、実習を行う。